Uploaded by IAEME PUBLICATION

LOAD BALANCING AND ENERGY EFFICIENCY IN WSN BY CLUSTER JOINING METHOD

advertisement
International Journal of Advanced Research in Engineering and Technology (IJARET)
Volume 10, Issue 2, March-April 2019, pp. 124-134. Article ID: IJARET_10_02_014
Available online at http://www.iaeme.com/IJARET/issues.asp?JType=IJARET&VType=10&IType=02
ISSN Print: 0976-6480 and ISSN Online: 0976-6499
© IAEME Publication
LOAD BALANCING AND ENERGY EFFICIENCY
IN WSN BY CLUSTER JOINING METHOD
Dr Syeda Gauhar Fatima
Professor ECE dept, Deccan College of Engineering and Technology,
Darussalam, Hyderabad.
Syeda Kausar Fatima
Associate Professor, Shadan College of Engineering and Technology, Hyderabad.
Dr K.Anitha Sheela
Professor ECE dept, JNTUH, Kukatpally, Hyderabad.
Syed Shaker Hussaini
Student ECE dept, Deccan College of Engineering and Technology, Darussalam, Hyderabad
ABTRACT
In any WSN life of network is depending on life of sensor node. Thus, proper load
balancing is very useful for improving life of network. The tree-based routing protocols
like GSTEB used dynamic tree structures for routing without any formation of
collections. In cases of larger networks, the scheme is not always feasible. In this
proposed work cluster-based routing method is used. Cluster head is selected such that
it should be close to the base station and should have maximum residential energy than
other nodes selected for cluster formation. Size of cluster is controlled by using locationbased cluster joining method such that nodes selects their nearest collection head based
on the signal strength from cluster head and distance between node and cluster head.
Nodes connect to head having the highest signal strength and closest to the base station,
this minimizes size of cluster and reduces extra energy consumption. In addition to this
cluster formation process starts only after availability of data due to an event. So
proposed protocol performs better than existing tree based protocols like GSTEB in
terms of energy efficiency.
Keywords: Wireless Sensor Network, Energy efficiency, load balancing, clustering,
routing.
Cite this Article: Dr Syeda Gauhar Fatima, Syeda Kausar Fatima, Dr K.Anitha Sheela
and Syed Shaker Hussaini, Load Balancing and Energy Efficiency in WSN by Cluster
Joining Method, International Journal of Advanced Research in Engineering and
Technology, 10(2), 2019, pp. 124-134.
http://www.iaeme.com/IJARET/issues.asp?JType=IJARET&VType=10&IType=2
http://www.iaeme.com/IJARET/index.asp
124
editor@iaeme.com
Load Balancing and Energy Efficiency in WSN by Cluster Joining Method
1. INTRODUCTION
Wireless sensor network is made up of number of sensor nodes mostly deployed in large
quantity due to their energy limitation. The way by using which data gathering is performed is
very important to any WSN application. Information collection and routing based on clusters
are widely used method because it provides better energy efficiency .Load balancing clusters
are formed by considering their position and considering their residential energy. One node is
selected as a head node which collects all the data from group members, aggregates them and
finally forwards it by using routing algorithm .during each round role of cluster head changes
it produces better load balancing .Clustering avoids communication of same data from multiple
path and minimizes energy consumption. With energy other factors related to the QoS must be
considered for better performance of the system .One of the important QoS limitation is delay
.End to End delay of WSN should be minimum[ 1 ].It is possible to achieve this objective by
using cluster based routing .We can save energy either at network layer by using proper routing
algorithm and at MAC layer by using proper MAC protocol .By selecting proper routing
protocol and proper MAC protocol we can also control delay in any WSN application. Delay
should be important design issue in particular types of applications such as military applications
,earthquake monitoring[ 3 ] . Clustering is useful to reduce communication distance of nodes
to reduce energy required for communication of data[4].In any WSN hierarchical routing
mechanism is useful to avoid redundant data communication. We can also conserve energy by
adopting proper
Duty cycle by keeping nodes active which have data to sense and to transmit and keeping
all remaining nodes to sleep state who don’t have any data to transmit .Combination of these
two methods can network life time and overall performance[6].
Use of mobile sink is good idea to balance network load properly in order to improve
lifespan of system. Tree based routing is also used for load balancing. Work GSTEB [09] has
provided this type of protocol and in work [10] has provided improved version of this protocol.
In proposed work cluster head is selected close to the base station. In addition we have tried
to control unnecessary size of cluster because redundant extra size of cluster consumes extra
energy. To handle this issue during formation of group member nodes selects their cluster head
based on distance and signal strength coming from cluster heads. TDMA is also used as medium
access protocol to improve performance.
If cluster formation starts only after availability of data due to event, this minimizes power
required to construct redundant cluster formation which doesn’t have any information[11].We
used this method for more energy efficiency .
Section two is provided with related work in the same area. Third section provides
architecture of proposed work and detailed explanation of protocol. Fourth section provides
performance analysis and comparison graph with existing system.
2. RELETED WORK
In many WSN applications with life of network some quality of service must be provided. Some
research work on WSN is only considering lifetime of network but QoS should be considered
mostly in case of applications such as military surveillance and battleground monitoring in case
of these applications end to end delay should be minimum .Author Min Yao etal. of work[ 1
]has tried to handle this issue .They have proposed model which can consider both problem of
lifetime of network and QoS requirements mostly the delay between source and sink and data
loss during communication process .Their results show that the work improves network lifetime
with maintaining QoS requirements [ 1 ].
http://www.iaeme.com/IJARET/index.asp
125
editor@iaeme.com
Dr Syeda Gauhar Fatima, Syeda Kausar Fatima, Dr K.Anitha Sheela and Syed Shaker Hussaini
Periodically turning off the radio of sensor node is good solution to energy consumption but
this may increase end to end delay in the WSN .Here number of packets to the sensor node are
controlled to minimize energy consumption and end to end delay of the system .author Yun Li
etal. Of [2] used dynamic programming approach to create proper communication policy to
improve performance of the system .their simulation results shows that their work is 50% better
than existing IEEE802.15.4 work.
In case of LEACH when node has no data in its TDMA slot its TDMA slots will be wasted
.Work[3] op leach handles this by conveying TDMA slots of nodes having no data to send by
conveying its TDMA slot to next node to avoid wasting of TDMA slot .This work is useful for
both energy efficiency and delay minimization. [ 3 ].
Another important work is provided in paper [4] This work used concept of clusters similar
to LEACH for routing and load balancing but they have also provided solution on security
threats on cluster heads such as gray hole attack. Their results shows that their system is better
than LEACH because their proposed work also addresses problem of gray hole attack. This is
the attack in which any malicious sensor device can send fake data to the cluster head to engage
sensor node in an unnecessary activity .This will increase energy consumption and delay in the
system [4].
t is very important to solve problem of global clock synchronization .author Supantha Das
etal. Of work [5] tried this by avoiding use of clock synchronization instead they have provided
new technique of coloring. Network is organized in tree structure .Parallel communication is
achieved for better performance. Their results shows that their system works better as compared
to other related work in terms of both energy consumption and End to End delay .Instead of
using global time they have used local time synchronization .Inter level and intra level coloring
method is used to improve parallelism and to avoid impact [5]. Combined approach of energy
efficient scheduling and hierarchical routing is provided by author of work [6].First they apply
power aware scheduling by activating only those nodes involved in data processing and
communication and remaining nodes are in sleep mode .After that hierarchical routing is used
for power saving. This combined approach works better in terms of energy efficiency. Their
simulation results shows that their work is better than existing systems such as LEACH, PASC,
M GEAR systems. [6].
Use of mobile sink is good idea for energy efficient load balancing ,such type of work is
presented by author Wen-Hwa Liao etal. in paper [7].They have categorized nodes in to
rendezvous nodes and non-rendezvous nodes .Here only rendezvous nodes can communicate
with the mobile sink and non-rendezvous nodes can send their data to the rendezvous nodes
.Here mobile sink can travel to the rendezvous nodes in the area and collect desired data in the
energy efficient way and this works well for improving network life time by reducing distance
between sink and rendezvous nodes[7].
In any WSN when more than one node tries to transmit data at the same time using single
communication medium then it will create contention for accessing medium and impact of data
.If impact is more than particular limit this will consume extra energy and delay in the system
due to recommunication of data and reduces QoS requirements of the system .If contention is
due to correlated traffic there is possibility of more contention so author Ashutosh Bhatia etal.of
[ 8 ] tried to solve this problem .TDMA is good solution for this but according to [8 ]if TDMA
schedule is not generated within short periods of time then it will be useless because correlated
contention presents only for short duration of time so author has proposed RD-TDMA
algorithm which generates feasible TDMA schedule in very short period of time[8 ] .
Tree based routing is very useful for load balancing. Work [09] by author Zhao Han etal.has
used tree based routing for load balancing . dynamically created tree is used for communication
of data from sensor node to root . Here tree will be created in minimum time and load is properly
http://www.iaeme.com/IJARET/index.asp
126
editor@iaeme.com
Load Balancing and Energy Efficiency in WSN by Cluster Joining Method
balanced using tree structure .Work [09]is best but problem with [09] is that load is not
distributed in node centric way and base station is involved in tree construction so lot of data
communication between base station and remaining nodes will increase delay. To solve this
problem work number [10] by M.Sengaliappanetal. has provided improved version of this tree
based routing .Author has used cluster based routing for node centric load balancing .In this
work base station is very less involved in actual data communication so delay in the system is
also minimized[10]. Cluster formation only after event detection provides better performing
network because it reduces unnecessary energy consumption .Work [11]is same type of work
.They have implemented their work using MATLAB simulator. Their results are better than
existing cluster based applications SCHS and EESH.
Use of multiple sink is another way for load balancing and for timely delivery of data,author
of work [12] has tried to solve problem of load balancing by using multiple sinks instead of
using single sink. They have provided efficient task allocation method in large scale WSN.Their
results are better in terms of packet delivery ratio, speed of data aggregation and time required
for processing and communication of data. Multiple sinks used are mobile in nature it improves
total performance of the system [12].
Cross layer optimization is better way for energy efficiency and load balancing ,work[13]is
based on this approach .With collecting they have also used cross layer approach to improve
life of the network .This work focuses on all three main layers at which we can control power
consumption .This layers are physical layer ,MAC layer and routing layer .At physical layer
they control communication power .At Medium Access Layer they have used S-MAC duty
cycling and at network layer they have implemented energy efficient routing method This
network has capability of dynamically reconfiguration based on current condition of the
network and network topology change. [13].Similar work is provided by author of [14].Node
sleeping is good for energy efficiency but it increases unnecessary end to end delay in the WSN
,so this author has tried to address this issue by adopting new routing method .Wake up rate is
provided with each node ,using this wake up rate relay node is selected for routing purpose. It
is very useful to control delay in the network. This
works well than heuristic
methods[14].Optimal woke up rate is provided because large wake up rate increases extra
energy consumption. Data packets are created and forwarded only after detection of event and
this avoids unnecessary energy consumption of the sensor node.
Node close to the base station or sink will consume energy in fast way. This will create
holes in the network and this will result in early death of network. To avoid this author of
work[15]provided concept of multiple sink. When multiple sinks are used then total load is
distributed properly. Instead of creating single routing tree they have created multiple routing
trees towards each sink in the network. This reduces extra time and energy consumption[15].
Author of work [16] considers both these performance aspects and provided tradeoff
between End to End delay and energy consumption .New algorithms are provided for inter
cluster routing and for intra cluster routing .Two functions are created for both for optimization
purpose .Clusters are created in distributed way to evenly distribute network load on all the
cluster nodes .Cluster heads are selected by considering both factors energy and delay. They
have provided results in two forms theoretical and simulation, both results are same up to some
percentage [16].
Congestion free and free wireless sensor network provides better output in terms of energy
,delay and minimum packet drop. Author of work[17] has considered and implemented this
issue by constructing congestion free routing tree .They have proposed CATopology i.e.
Congestion Avoidance Topology by using two methods known as K-Map and K-Graph.
Routing on this multi hop routing tree dose not allows more than one packets to reach at the
same point in the network at the same time period .This improves energy efficiency, minimizes
http://www.iaeme.com/IJARET/index.asp
127
editor@iaeme.com
Dr Syeda Gauhar Fatima, Syeda Kausar Fatima, Dr K.Anitha Sheela and Syed Shaker Hussaini
end to end delay and also minimizes number of packet drops in the network. They have
simulated their work by using OPNET network simulator .Their results are better than similar
work[17]. Author of work[18] provides method of cluster formation in distributed way. They
have tried to distribute network load on each and every node in the network. They have
addressed problem of scalability and energy efficiency. TDMA is also used for performance
improvement and sleep wakeup method is used for more energy saving .For selecting cluster
head both distance and residential energy of node is considered [18].
Work [19] has provided generic algorithm for load balancing purpose .They have used
different strategy for generation of initial population to generate chromosomes which are valid.
For proper load balancing they have used fitness function to balance network load properly.
Each cluster member node can communicate with cluster head node by using single hope or
using multi hope communication. According to the requirement. If node is very close to the
head it can communicate directly and if node is not close to cluster head then it can use malty
hope path for communication purpose. The work is better in terms of energy efficiency[19].
GSTEB is a tree based routing protocol. In GSTEB routing tree is generated dynamically
in each round. Here round means time from sensing data up to the time at which data is
delivered to the base station. Work[20]is based on improvement in tree based routing protocol
.Author of work[20] has added concept of fuzzy rules for creation of clusters and for selecting
proper routing path. Use of fuzzy rule has shown better performance in terms of energy
efficiency and delay minimization in the WSN. Network simulator 2 is used for implementation
purpose. Results show that system performs better than existing HEED and GSTEB protocols
[20].
Author of [21] has divided sensing area into four parts. Each part is termed as quadrant.
Parent or relay node is selected by considering distance between nodes and residential energy
of node ,this makes better relay node selection for minimum energy consumption. TDMA is
used for resource allocation purpose .Each cluster head performs TDMA schedule for each node
inside the cluster and broad cast this schedule to all cluster members. All cluster members
follows this TDMA schedule for data communication purpose. They have implemented their
work by using MATLAB simulator and results are better than existing NRLM and LEACH
protocols [21].
3. IMPLEMENTATION DETAILS
We assumed system model similar to the GSTEB. Sensor nodes are deployed in square area
randomly. Only single base station is provided and which is far away from the sensing area
.Any node can communicate directly with the base station if it is closer to it by changing its
current remaining energy. All sensor nodes have their fixed position .That is it is assumed that
they cannot change their position .Base station is also stationary but it is assumed that it is no
energy constrained
http://www.iaeme.com/IJARET/index.asp
128
editor@iaeme.com
Load Balancing and Energy Efficiency in WSN by Cluster Joining Method
3.1. System architecture of proposed work
Figure 1 Closest cluster head to the base station with maximum energy level.
Above figure 1 shows architecture for proposed protocol. Proposed work has four phases
which are described in following section.
3.2. Phases involved in proposed approach
3.2.1. Phase 1. Cluster formation phase
Cluster formation process starts with a node which listens some event related to the availability
of data, we can say it as a seed node. Now other nodes which are in the communication range
of seed node accept the cluster formation request from seed node and joins this node for cluster
formation process .If two nodes listen event at the same time then node having maximum energy
is selected as seed node .If two nodes listen event at the same time but at large distance from
each other such that they are not within communication range of each other ,in this case both
are considered as seeds for two different clusters it is shown in fig 2.
Base is indicated by green circle. Here in the area two seed nodes are selected and they are
indicated by brown colored circles. Communication range of each seed is considered as 250
meter, here both seeds are not within communication range of each other ,so both are
considered as two separate seeds for two separate cluster formation. Violet color nodes are
within communication range of seed1 and magenta color nodes are within communication range
of seed2.
Algorithm for seed node selection
Step 1: If seed1 and seed2 are within communication range of each other and energy of
seed1 is greater than seed2 then seed1 is selected as seed node
Step2: If seed1 and seed2 are not within communication range of each other then both are
considered as separate seed nodes for two deferent cluster formations
Step3: Each selected seed sends cluster formation request to its neighbors who are within
its communication rage.
Step4: Nodes within communication range of seed are selected for cluster formation process
We can calculate energy level of any node by using following equation.
E(i)= Residual Energy(i)/α
(1)
If there are N number of nodes then i indicates i th node. E= Current energy level.
α = smallest energy unit.
http://www.iaeme.com/IJARET/index.asp
129
editor@iaeme.com
Dr Syeda Gauhar Fatima, Syeda Kausar Fatima, Dr K.Anitha Sheela and Syed Shaker Hussaini
3.2.2. Phase 2. Cluster Head Selection Phase
In this phase cluster head is selected based on two criteria’s. first is distance between base
station and candidate nodes for cluster head selection and second criteria is energy level of
node. First criteria selects cluster head closest to the base station .Selecting closest cluster head
to the base station minimizes energy consumption for communication of data .This is useful
because energy required for communication of data is higher than energy required for receiving
the data and for processing the data. Based on these two criteria values for each node in the
cluster is calculated. Node having maximum value is selected as cluster head.
Figure 3 Cluster head selection
As shown in the fig 3 base stations are indicated by green color. Three cluster heads are
selected for three different cluster formations. Cluster heads are indicated by light orange color.
Members of cluster for head 1 are shown by dark orange color and members of cluster for head
2 are shown by magenta color and members of cluster for head 3 are shown by dark violet color.
Important point here is to note that all these cluster heads are close to the base station and having
high energy.
Algorithm for cluster head selection phase
Step1: For all nodes selected by each seed distance between nodes and base station is
calculated.
Step2: Node closest to the base station and having highest energy is selected as cluster head.
Step3: If there are two nodes having same highest energy and closest to the base station
then cluster head is selected
randomly
We know that the length of line in between any points P and Q is known as Euclidean
distance [12].We have used same formula to calculate distance between sensor nodes If p and
q are two nodes then we can calculate distance between them as follow [12] .
D(p,q)=
(2)
3.2.3. Phase 3. Location based cluster joining and controlling cluster size.
After selecting cluster head which is close to the base station and having maximum energy level
.Now it is time to control size of cluster .Because unnecessary extra size of cluster will increases
energy consumption .In this phase selected cluster head sends request message to join to the
remaining nodes in the cluster. Based on strength of message coming from all the nearest head
and based on distance between member nodes and CH, a node selects their cluster head. In this
way node selects CH based on physical position that is distance and using actual performance
of cluster head i.e. its signal strength. This controls size of cluster in effective way .According
to the architecture shown in figure 1 sensor node N1 and node N2 will not becoming part of
cluster of head C1 to control size of cluster.
http://www.iaeme.com/IJARET/index.asp
130
editor@iaeme.com
Load Balancing and Energy Efficiency in WSN by Cluster Joining Method
Figure 4 Members eliminated from cluster formation
The Fig 4 shows simulation results of controlling size of cluster. Here all the dark colored
nodes are eliminated from cluster formation to control unnecessary size of cluster. After
selecting cluster head nodes sends accept message to selected cluster head, with this message
nodes also sends control information such as residential energy to the respective CH. This
minimizes time required to construct routing tree and minimizes delay in the network for each
round.
Algorithm for cluster joining phase
Step 1: Each cluster head sends join request to its members.
Step 2: All members finds signal strength coming from cluster head.
Step 3: Nodes selects cluster head close to it and having maximum signal strength
Step 4: If signal strength is weak then node will not become member of cluster to reduce
size of cluster.
3.2.4. Phase 4 : Cluster tree creation and actual data communication.
Now each cluster head selects node closer to the base station by calculating weight of all
neighboring nodes from following weight vector.
W=W1, W2, W3,WN
Dijkstras shortest path algorithm is good for finding the shortest path but it only considers
distance while selecting the shortest distance .But in WSN with distance energy of node must
be considered while selecting parent node[11] Here weight for each node is calculated based
on distance and energy. Here optimum path is selected between cluster head and base station.
In this way from each cluster head optimum path is selected to the base station .it provides tree
like structure and we can say it as routing tree.
Finally Sensor nodes sense data and creates DATA packets and forwards this DATA packet
to cluster head only within its time slot provided by cluster head. Energy required for sensing
can be calculated by using following equation.
ES= Z * PSi * t.
(3)
Here PSi is the size of sensed data.t is the time required for sensing .and Z is the constant.
Now cluster head performs processing such as data aggregation and forwards data without
any redundant information by using routing tree created based on weight value .Energy required
for communication and reception of k bit of data over the distance d can be calculated as
Energy required for communication of data
ET (k, d) = (Eelec × k) + (Ɛamp × k × d2)
(4)
Energy required for reseption of data
http://www.iaeme.com/IJARET/index.asp
131
editor@iaeme.com
Dr Syeda Gauhar Fatima, Syeda Kausar Fatima, Dr K.Anitha Sheela and Syed Shaker Hussaini
ER ( k) =(Eelec × k )
(5)
Here Energy consumed by circuit of transmitter or receiver is denoted by Eelec and `Energy
consumed by amplifier denoted by Ɛamp.
4. SIMULATION AND RESULTS
Protocol provided in this work performs better load balancing than existing tree based routing
protocol in wireless sensor network application. By using this new protocol we can avoid early
death of nodes close to the base station. This protocol distributes load on the network evenly
and provides proper load balancing and energy efficiency to improve life of WSN.Table2 shows
parameters used during implementation of the system.
Table 1 Simulation Parameters
100 to 400
0.25 J /node
Simulation
Parameters
Packet size
Traffic Mode
2000 bits
Poisson
50n J/ bit
Area of sensing field
100m X 100m
100 pj/bit/m2
Base station location
(50m,150m
)
Simulation Parameters
Value
No of Sensor Nodes
Initial Energy of Each Node
Electronics Energy
Consumption
Amplifier transmitting
energy
Energy consumption for
aggregation
Value
5nJ/bit/signal
Our NS2 simulation results shows that our system has better performance than existing tree
based routing protocols such as GSTEB protocol in terms of energy efficient routing and load
balancing. We can see it by using following graphs. Here LCJM is a short name given to our
protocol i.e. Location Based Cluster Joining Method fig 5 shows comparison for total energy
consumption per round, from the figure it is clear that total energy consumption of proposed
work is less as compared to GSTEB and figure 6 shows graph for number of dead nodes per
round. Here after 100 rounds GSTEB has 9 dead nodes but proposed work has only 3 dead
nodes. Graphs using simulation result shows that in same network conditions and by using
same parameters used by GSTEB proposed work performs better
Figure 5 Total energy consumption per round
http://www.iaeme.com/IJARET/index.asp
132
editor@iaeme.com
Load Balancing and Energy Efficiency in WSN by Cluster Joining Method
Figure 6 Number of rounds vs. number of dead nodes
5. CONCLUSION
This protocol is different than previous tree based and simple cluster based systems because
cluster heads are selected close to the base station to minimize energy consumption due to
communication of data from cluster head to the base station .Location based cluster joining
method is used to control unnecessary size of cluster to minimize energy consumption. Process
of cluster formation starts only after availability of data So these feature provides better load
balancing for energy efficient routing to improve life of the network
ACKNOWLEDGMENT
This work is supported by K.J.College of Engineering and Management Research .I would like
to thank my project guide and P.G coordinator for proper guidelines. I would also like to thank
all other staff members of my college for their support.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
Min Yao, Chuang Lin, Yuan Tian “Energy and Delay Minimization in Cluster-Based
Wireless Sensor Networks”, International Conference on Green Computing and
Communications, IEEE 2012.
Yun Li, KokKeong Chai, Yue Chen “Optimal Communication Policy with Joint
Minimization on Energy Consumption and Delay for Wireless Sensor Networks”,
ICCT,IEEE 2013
NabeelaAnsari , Raj Kumar Paul,“ Modified Leach in Wireless Sensor Network”, IOSR
Journal of Computer Engineering (IOSR-JCE), Nov – Dec. 2014.
S. Anbuchelian; K. Selvamani; A. Chandarasekar, “An energy efficient multipath routing
scheme by preventing threats in Wireless Sensor Networks”, Electrical and Computer
Engineering (CCECE), IEEE 2014
Supantha Das, Indrajit Banerjee, MainakChatterjee and TuhinaSamanta ,“Performance
analysis of TDMA Based Data Communication in WSN”,IEEE 2014
CherifaBoucetta; HanenIdoudi; Leila AzouzSaidane, “Ant Colony Optimization based
hierarchical data dissemination in WSN”, International Wireless Communications and
Mobile Computing Conference (IWCMC).IEEE 2015
Wen-Hwa Liao , Yi-An Yen, and Ssu-Chi Kuai “An Efficient Load Balance Data Collection
Scheme in Wireless Sensor Networks”,IEEE 2015
Ashutosh Bhatia , R. C. Hansdah ,“RD-TDMA: A Randomized and Distributed TDMA
Scheduling for Correlated Contention in WSNs”,Advanced Information Networking and
Applications Workshops (WAINA) Victoria BC May 2014
http://www.iaeme.com/IJARET/index.asp
133
editor@iaeme.com
Dr Syeda Gauhar Fatima, Syeda Kausar Fatima, Dr K.Anitha Sheela and Syed Shaker Hussaini
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
Zhao Han, Jie Wu, Member, IEEE, Jie Zhang, Liefeng Liu, and KaiyunTian,“A General
Self-Organized Tree-Based Energy-Balance Routing Protocol for Wireless Sensor
Network”, Nuclear Science, IEEE Transactions ,IEE April 2014
M.SENGALIAPPAN, Dr.A.marimuthu “improved general self-organized tree-based
routing protocol for wireless sensor network”, journal of Theoretical and Applied
Information Technology,OCTOMBER 2014.
Sohini Roy, Ayan Kumar Das, “Energy Efficient Cluster Based Routing Protocol
(EECBRP) for Wireless Sensor Network “,IEEE 2014.
S. PrabhavathiDr.A. AnandaRaoDr.A.Subramanyam ,“Globular Topology of Large Scale
WSN for Efficient Load Balancing Using Multiple Sink Node”, Computing,
Communications and Networking Technologies (ICCCNT) ,Fourth International
Conference IEEE,2013
Zouhair A. Sadouq; Marouane El Mabrouk; Mohamed Essaaidi, “Conserving energy in
WSN through clustering and power control”, Third IEEE International Colloquium in
Information Science and Technology (CIST),2014.
Priyanka M. Lokhande* and A. P. Thakare “Maximization of Lifetime and Minimization
of Delay for Performance Enhancement of WSN”, nternational Journal of Business
Research and Development Vol. 4 No. 3, pp. 9-16 ,2015
PunyashaChatterjee ,Nabanita Das “Multiple Sink Deployment in Multi-Hop Wireless
Sensor Networks to Enhance Lifetime”, Conference on Applications and Innovations in
Mobile Computing (AIMoC),IEE,2015.
Trong-Thua Huynh; Anh-Vu Dinh-Duc; Cong-Hung Tran, “Delay constrained energyefficient cluster-based multi-hop routing inwireless sensor networks”,Journal of
Communications and Networks Volume: 18, Issue: 4,IEEE 2016.
RanjanDasgupta; Ritwick Mukherjee; Amitava Gupta, “Congestion avoidance topology in
wireless sensor network using Karnaughmap”,Confarance on Applications and Innovations
in Mobile Computing (AIMoC),IEEE, 2015.
ChirihaneGherbi; ZiboudaAliouat; Mohammed Benmohammed, “Distributed energy
efficient adaptive clustering protocol with data gathering for large scale wireless sensor
networks”,conference on Programming and Systems (ISPS), IEEE,2015.
NabajyotiMazumdar , Han Om “An Energy efficient GA-based algorithm for clustering in
wireless sensor networks”, International Conference on Emerging Trends in Engineering,
Technology and Science (ICETETS), ,IEEE, Feb 2016.
KaushikGotefode , KishorKolhe “Energy Efficiency in Wireless Sensor Network using
Fuzzy rule and Tree Based Routing Protocol”, International Conference on Energy Systems
and Applications, Pune,IEEE 30 oct to 1nov 2015.
VipinPariharParulKansal,” Quadrant based routing protocol For improving network
lifetime For WSN”India Conference(INDICON),IEEE, 17 to 20 December 2015
ABOUT AUTHOR
First Author Completing his Master of Engineering at K.J College of engineering and
management research, Pune University. He has completed Bachelor of Engineering from
P.V.P.I.T, Shivaji University, Kolhapur.
Second Author Working as assistant professor and head of department at Computer
Engineering department at KJCOEMR Pune, savitribaiPhule University. He has 10 years
teaching experience. He has completed M Tech in computer science and engineering and PhDPursuing at sathyabama university, Chennai.
http://www.iaeme.com/IJARET/index.asp
134
editor@iaeme.com
Download